Artificial Intelligence (AI) is an emerging technology with a wide range of applications for government, non-government organisations and industry. The NSW Government will leverage the significant potential of AI to deliver better customer outcomes, and make better decisions, for the people of NSW.
The NSW Government will ensure that AI solutions used by government are trusted by the public, meet the highest ethical and assurance standards, are clearly focussed on customer needs, and carefully manage potential risks.
AI will not be used where there is not a clear use case for doing so, or where its use might pose risks in relation to data, privacy or assurance.
Accordingly, the NSW Government has developed ethical principles for best practice use of AI that are focused on trust and transparency, high quality outcomes and customer benefit.
This approach is informed by the extensive work on AI in other jurisdictions within Australia and internationally such as the European Union, OECD, industry and non-government organisations. It has also been informed by detailed consultation with the community, non-government organisations, government agencies, academia and with industry.
The principles are focused on better customer service and better informed decision-making.
There are many definitions of AI in academia, industry and government. AI can refer to robotics, machine learning and the simulation of human intelligence by machines and computer systems. The overall goal of artificial intelligence is to create technology that allows computers and machines to function in an intelligent manner.
This policy defines AI as intelligent technology, programs and the use of advanced computing algorithms that can augment decision making by identifying meaningful patterns in data. AI in this context should aim to help government agencies cut costs, free up labour hours for more critical tasks, and deliver better, more targeted services.
AI has the capacity to quickly process large amounts of data and provide data-driven insights, supporting problem solving and decision making to maximise efficiency and service delivery.
There are a number of examples of AI-use in NSW Government:
- Transport for NSW uses sensors to gather data, and AI and machine learning to analyse that data, to assess the risk of a failure, anticipate required maintenance and proactively schedule downtime required for that maintenance, increasing safety and reliability.
- NSW Health is using algorithms, trained on data from deidentified medical records, for early detection and treatment of sepsis in emergency department waiting rooms, saving patient lives and improving service levels.
The NSW Government is focused on improving customer service and community outcomes. The Premier’s Priorities are a commitment to significantly improving the quality of life for our communities:
- Lifting education standards – lifting academic performance and improving outcomes for all students, with a particular focus on Aboriginal students
- Keeping children safe – protecting the vulnerable and increasing permanency in out of home care
- Breaking the cycle – reducing domestic violence, recidivism in the prison population and homelessness
- Improving the health system – improved service levels in hospitals and outpatient and community care, and reduced suicide deaths
- Better environment – greener public spaces and a greener city
- Better customer service – government made easy and a world class public service.
These are complex issues requiring collaboration across agencies to understand risks, develop intervention strategies and design policy and service delivery.
AI is a powerful tool that has the potential to help us in these priority work areas, and other complex public policy areas that require analysis and solution development.
The AI Policy and Assurance Framework are designed to:
- demystify AI for NSW Government project managers
- authorise the use of AI while providing guidance on best practice and risks
- encourage innovative approaches to service delivery and decision-making
- build understanding and capability across government.
The potential benefits of AI are significant, but its use needs to be managed carefully to ensure risks are controlled and unintended outcomes avoided. The AI Assurance Framework has been designed to do this